We heard in the grapevine that BGI assembled a very complex
highly insanely polymorphic) genome of an eukaryotic organism.
The paper is close to being accepted, and we hope to learn some new techniques
on genome assembly from it.
We have not commented much on SOAPdenovo, the de Bruijn graph-based genome assembly program from BGI. It is supposedly one of the best assemblers out there based on past Assemblathon results. However, the source code was not released and therefore it was difficult for us to comment much on the algorithm. All that is going to change soon, based on these correspondences sent to BGI-SOAP mailing list from Ruibang Luo of BGI.
Wed, Jun 13, 2012 at 7:44 PM
And we will release SOAPdenovo2 soon, we may put it on SourceForge, where the interactions between developer and users can be searched by other users with similar problems.
Sun, Jun 17, 2012 at 8:03 PM
It’s about to release in next month.
For you convenience, you can join out testing to use SOAPdenovo2 in advance to the formal publish.
If you do not know much about the operations at BGI, here is a nice overview from Nature news.
Some interesting factoids:
1. Although BGI started in Beijing, it now moved to Shenzhen in south China. How far are those two cities? You can imagine Caltech moving to Houston.
2. BGI operates by taking a 10-billion-renminbi loan (US $1.57B) from the China Development Bank. So, researchers are highly indebted to the bank. That is quite different from US model, where professors’ salaries come from students being put on lifelong debt.